Interesting effect with new side-to-side transfer pumps...

[IDBoating]

Today I installed my side-to-side transfer pumps. I'm using two Rule 405FC's running between the aft sacs to 1) double the speed of fill and drain, and 2) double the speed of switching from normal to goofy side. Each pump takes its intake from the bottom of one sac and fills into a top port on the opposite sac.

Everything worked great right from the start. Filling and draining are much faster, as is switching sides. However, I noticed that I was losing water out of the port (normal) side sac. Maybe I have a leak, I thought. After we switched sides, I found I was losing water out of the starboard sac too! Odds of matching leaks on both sides: Almost zero.

A bit of investigation revealed what is happening: When I drain one side, it is actually suctioning water through the fill hose from the other sac. Thus it is not only draining the intended sac, but the sac on the other side too - through the transfer hose that intentionally moves water that direction when the associated transfer pump is turned on.

I can't fix this with a check valve because the suctioned water is moving in the same direction as the desired water when the pump is on. At the moment the only idea I have is to make sure there is an air pocket in the top of each sac so when the transfer pump isn't running there is a vacuum break to prevent the flow from starting.

Anyone else encounter this? Any better solutions?

[chpthril]

Im not sure I understand how the transfer pump is drawing water out of the very sac its filling, but the best pump for a transfer system is a reversible impeller pump. When off, it acts as its own check-valve.

[IDBoating]

Yes, an impeller pump might inherently fix this - but they're expensive and I'm cheap. In fact, I did think of that but then remembered your earlier comment that sometimes pressure overcomes the impeller and permits flow even when the pump is off, so I don't know if it's a guaranteed fix or not.

To be more detailed: Let's say I'm switching sides from normal (port) to goofy (starboard). I turn on the port drain pump, the starboard fill pump, and the port-to-starboard transfer pump. Because the pumps are differing sizes the starboard sac fills before the port sac empties, so I turn off the starboard fill pump and the port-to-starboard transfer pump.

Now the only pump running is the port drain pump. What happens is that as the port drain pump sucks water out of the port sac, it generates enough of a vacuum to also suck water through the starboard-to-port transfer pump and hose. Over a few minutes it will drop the top of the starboard sac by several inches. You can see the water flowing through the hose even though the transfer pump is not running.

So in essence each sac has a straw through which it can suck water from the opposite sac. I ran the transfer hoses from the bottom of the source sac to the top of the destination sac to prevent gravity-driven transfers, but apparently that's not enough to stop the drain pumps from drawing enough of a vacuum to pull water through the hose.

One test will be routing the transfer hoses a bit higher so they loop up and then drop down to the top of the sacs.

Comments and brainstorming is welcomed... thanks!

[mls6722]

you can - 1 - put a gate valve in the transfer lines to isolate the tanks, 2 - refill the sac after turning off the opposite drain pump.

[IDBoating]

Thank you for responding!

you can - 1 - put a gate valve in the transfer lines to isolate the tanks

Nope, because the undesired water flow direction is the same as the desired flow when the pump is running. Installing a valve in the line will block the water I WANT to flow.

EDIT: Unless you mean an electrically actuated valve, in which case Yes that would work but I'd have to really think hard about putting a valve in there. I'm no fan of valves in ballast systems.

2 - refill the sac after turning off the opposite drain pump.

That's what I did yesterday to keep things topped off, but it defeats one of the reasons I did this - which was to speed up changes in ballasting. If I have to sequence the pumps instead of running them at the same time, I won't get the speedup I should get from having two more pumps onboard.

I appreciate the response, please keep the ideas coming. There's a solution out there....

[chpthril]

So the port drain pump is drawing water through the transfer pump that is connected to the bottom of the starboard sac and used to move water from the starboard sac to the port sac and vis versa????

If so, then a manual or electric gate or ball valve will fix it and a check-valve will not.

In fact, I did think of that but then remembered your earlier comment that sometimes pressure overcomes the impeller and permits flow even when the pump is off, so I don't know if it's a guaranteed fix or not

yes, I have stated that, but in a different context. With a forward facing scoop, water can be pushed by the impeller. With a reverse facing scoop, a vacuum can be drawn. As a transfer pump, you should not experience this. The water pressure created by gravity should not push past the impeller. If this is still a concern, mount the pump high. This reduces the water pressure on the impeller.

[shawndoggy]

I'd just try an extra long hose run from side to side with a loop in the middle tucked up as high under the gunnel as you can. If that doesn't work then you need to go the vacuum breaker route like malibu uses



(at which point after two pumps, two hose runs, two sets of fittings, and two vacuum breakers... that "expensive" reversible pump doesn't look so bad).

[mls6722]

Thank you for responding!



Nope, because the undesired water flow direction is the same as the desired flow when the pump is running. Installing a valve in the line will block the water I WANT to flow.

EDIT: Unless you mean an electrically actuated valve, in which case Yes that would work but I'd have to really think hard about putting a valve in there. I'm no fan of valves in ballast systems.



That's what I did yesterday to keep things topped off, but it defeats one of the reasons I did this - which was to speed up changes in ballasting. If I have to sequence the pumps instead of running them at the same time, I won't get the speedup I should get from having two more pumps onboard.

I appreciate the response, please keep the ideas coming. There's a solution out there....

Yes, I was thinking of an actuated valve that closes when no cross over is occurring ($$$)

Another idea - put a pair of Y's in the crossover line with a check valve in each leg (between the Y's) with one in each direction. These check valves have a positive cracking pressure that should prevent the unwanted flow yet still flow when the pump is running.

[TigeTrojan]

How about a Y with a one way valve that will allow air in but wont let water out. When there is positive pressure in the line because water is intentionally being pumped through, the vent is closed by the water pressure. When there is unwanted negative pressure, the vent allows air in and stops the siphon before it starts.

[shawndoggy]

How about a Y with a one way valve that will allow air in but wont let water out. When there is positive pressure in the line because water is intentionally being pumped through, the vent is closed by the water pressure. When there is unwanted negative pressure, the vent allows air in and stops the siphon before it starts.

I think that's basically what I posted the picture of....

[TigeTrojan]

Ya I guess so, I thought you were saying that a Malibu vacuum breaker was expensive, so I was just giving the bubba version. After rereading, I got you. Same thing.

[chpthril]

Yeah thats pretty much what a vented-loop does.

[IDBoating]

I'd just try an extra long hose run from side to side with a loop in the middle tucked up as high under the gunnel as you can.

Yep, as I stated above:

One test will be routing the transfer hoses a bit higher so they loop up and then drop down to the top of the sacs.

I already do this with my fill hoses to prevent syphoning. I'm not doing it yet with the transfer hoses but left enough length in case I needed to. That's going to be my first test.

If that doesn't work then you need to go the vacuum breaker route like malibu uses

That works if it's a syphon problem, but that's not what is happening here. I'm not sure it will help in this situation, since instead of syphoning I have an active drain pump drawing a vacuum that is using the transfer hose as a soda straw, literally sucking the water out of the opposite sac. However, if nothing else works I may try that. So far my system has not required any valves nor vacuum breaks and I'm hoping to keep it that way.

(Note to ChpThril: I still have the two unidirectional valves I purchased from you. They were going to be used for a different project but prototyping showed it wouldn't work, so now they're just stock on hand.)

[TigeTrojan]

That works if it's a syphon problem, but that's not what is happening here. I'm not sure it will help in this situation, since instead of syphoning I have an active drain pump drawing a vacuum that is using the transfer hose as a soda straw, literally sucking the water out of the opposite sac. However, if nothing else works I may try that. So far my system has not required any valves nor vacuum breaks and I'm hoping to keep it that way.

The vacuum that you are describing is what I was calling a siphon. No matter what you call it, the vent will stop it.

[IDBoating]

The problem has been fixed, with no extra parts required other than two tyraps.

I spent some time thinking about what I was seeing and the angle of the hull under various conditions. It occurred to me that once one side is full and the other side is empty, the crossover fill port on top of the empty sac can be below the waterline inside the full sac. Since the two sacs are connected via a hose that runs from the bottom of the full sac to the top of the empty one, the old adage of "water will seek a common level" kicks in and water will start to flow through the hose in an attempt to equalize the water levels in the two sacs.

This is possible because when the top of the empty sac settles down, the crossover fill fitting on its top ends up below the full sac's water level. It's not a syphon - it's just water in two connected containers trying to equalize itself.

The fix is easy: Just loop the two crossover hoses up in the locker such that their high points stay above the waterline in the opposite sac. The full sac will always be lower (because of hull tilt), so if the opposite side's transfer hose is near the top of its locker it will be helped by the hull tilt and will be well above the waterline in the opposite side's full sac.

I accomplished this with two tyraps of the style that have the screw holes in their ends. I loosely tyraped around the hose, then put the screw hole over the engine divider's retaining bolt at the top of the locker and screwed the wingnut on top to hold everything. Nice, simple, ultra-cheap, and reliable.

We went out for several hours today and had zero loss, leakage, or unintended transfers of water. Filled sacs stayed topped off the entire time with no extra effort whatsoever.

Best of all, our fill and drain times are now about half what they used to be. We have about 2700 GPH of pumping capacity filling and draining our choice of the aft sacs, with no extra holes in the hull and just a single new switch. We also have about that flow rate when transferring from side to side.

Total cost: Roughly $150 for two Rule 405FC pumps, some hose, and some fittings. Oh yeah, and two tyraps. WELL worth it for all the speed increases.

Hope this helps someone!